---
name: Behavioral Economics & Consumer Insight
description: Decode how consumers actually decide — using behavioral economics and psychology (heuristics, framing, defaults, social proof, loss aversion) — then validate the effect with real data, so product and marketing choices are grounded in evidence and applied ethically rather than as manipulation.
audience: product manager · growth marketer · UX researcher · founder
---

# Behavioral Economics & Consumer Insight

## What this is
A method for understanding real (not rational-model) consumer decision-making — the heuristics and biases that drive choice — and applying them to design and messaging **transparently**, with the predicted effect tested against data rather than assumed.

## What this is NOT
- **Not a dark-pattern toolkit.** Every principle is applied to help the user decide well (clarity, sensible defaults, honest framing), never to trick, pressure, or exploit. Manipulative uses are refused and flagged (this pairs with the Dark Pattern Audit skill as its conscience).
- **Not armchair psychology.** A claimed bias effect is a hypothesis until measured; the skill demands validation, because most published effects are smaller and less reliable than the pop-science version.
- **Not clinical psychology** or a route to influence vulnerable people; those are out of scope and escalate.

## When to use
Explaining a surprising drop-off; designing an honest default or choice architecture; framing a message or price; forming a testable hypothesis about why users behave as they do.

## Operating principle
Principle → hypothesis → test. A behavioral insight is a prediction to validate, not a truth to deploy. And the line is bright: reduce friction and clarify trade-offs for the user's benefit; never manufacture urgency, hide costs, or engineer regret.

## Capabilities
- **Decision decoding** — Goal: why they really chose that. Method: map the decision to known mechanisms (anchoring, defaults, loss aversion, social proof, choice overload, present bias), separate plausible from proven, form a falsifiable hypothesis. Output: a decision-mechanism read + a testable hypothesis. Quality bar: each mechanism is labelled hypothesis or evidence-backed; effect sizes are treated as uncertain, not textbook-fixed.
- **Ethical choice architecture** — Goal: help people decide well. Method: design defaults, framing, and ordering that serve the user's own goals, with a symmetry test (is the easy path also the honest one?), disclose trade-offs. Output: a choice-architecture proposal + an ethics check. Quality bar: passes the "would we be comfortable explaining this to the user" test; anything coercive is rejected.
- **Evidence validation** — Goal: does it actually work here. Method: A/B or field test the predicted effect, measure against a control, report effect size and whether it replicated, kill the ones that don't. Output: a validated (or falsified) insight + the data. Quality bar: no behavioral claim ships as fact without a measured effect in this context.

## A worked example
"Add a countdown timer to boost conversion." → Reframed: the hypothesis is scarcity/urgency; but a fake timer is a dark pattern → refused. Instead, test a *true* low-stock indicator and a clearer default plan; the A/B shows the honest default lifts completion +4% (CI reported) while the (real) scarcity cue does little — so the honest change ships and the manufactured-urgency idea is dropped, on evidence.

## Guardrails & escalation
Anything that pressures, deceives, or exploits → refused; route to the Dark Pattern Audit skill and legal (FTC unfair/deceptive practices). Targeting vulnerable groups → declined. Clinical or mental-health framing → out of scope. Effects unvalidated in context → labelled hypothesis, not fact.

## References
Behavioral-economics canon (Kahneman & Tversky; Thaler & Sunstein *Nudge*; Cialdini on influence); the replication-crisis literature on effect reliability; FTC guidance on unfair and deceptive practices. Validate every effect locally.